posted 5 days ago on techcrunch
This next year will be one of the most important since Boston Dynamics was founded back in 1992. After changing hands from Google to Softbank, the robot maker is getting more aggressive about commercializing products, bringing Spot to market, while Handle waits in the wings. The news includes a new CEO, its first since founding. This week, Boston Dynamics is also making Spot’s SDK available to the public via Github. That will go live tomorrow. It’s a pretty big step for the company and its plans to grow its first commercially-available robot into a platform — something it’s talked up for a while now. VP Michael Perry offered up the following comment to TechCrunch, The SDK enables a broad range of developers and non-traditional roboticists to communicate with the robot and develop custom applications that enable Spot to do useful tasks across a wide range of industries. Developers will still need to become part of the Early Adopter Program to lease the robot to execute their code, but all interested parties will now be able to view the SDK and existing early adopters can open source their own code. With the SDK, developers in the Early Adopter Program can create custom methods of controlling the robot, integrate sensor information into data analysis tools, and design custom payloads which expand the capabilities of the base robot platform. One of our customers Holobuilder is using the SDK to integrate Spot into their existing app. With what they’ve developed, workers can use a phone to teach Spot to document a path around a construction site and then Spot will autonomously navigate that path and take 360 images that go right into their processing software. Other customers are exploring VR control, automated registration of laser-scanning, connecting Spot’s data to cloud work order services, using Edge computing to help Spot semantically understand its environment, and much more. Boston Dynamics has already showcased a number of potential applications for the robot on stage at TechCrunch’s annual Robotics+AI conference. Early uses include security and construction site monitoring. Spot’s ability to walk up and down stairs and open doors make it uniquely qualified among these sorts of robots. Another video, which featured Spot being used in state police drills, meanwhile, raised some concern with the ACLU. Of that, now-former CEO Marc Raibert told me, “There is a part of a humanity that loves to worry about robots taking over or being weaponized or something like that. We definitely want to counter that narrative. We’re not interested in weaponized robots. We’ve also gotten positive feedback from the fact that the police were using our robot to look at suspicious packages. There’s a real safety issue there and that it generated some additional interest with us as well. I mean, this isn’t really anything different than any new technology. There’s a wide variety of things it can be used for. We’re working to be responsible and trying find the good things that it could be used for.” The truth is that the nature of Boston Dynamics’ robots have — and probably always will — raise suspicions among an audience trained to be suspicious of large robots like Spot through generations of sci-fi stories. Certainly having it in the field with officers only contributed to such suspicions for many. Also true is that once Spot and the SDK are out in the world, BD will only have so much control over how such products are used in the world. One well-known early adopter is Adam Savage. The former Mythbuster got his hands on a Spot over the holidays and produced a video wherein he interacts with the robot like a kid on Christmas day. Understandable. I’ve controlled Spot myself and it’s pretty awesome once you get over the fleeting concern that you’re going to break a machine the size of a large dog that costs as much as a car. According to his video, Savage will be working with Spot for the next year.

Read More...
posted 5 days ago on techcrunch
After more than a quarter-century spent developing some of the industry’s most iconic and advanced machines, Boston Dynamics is a company in the midst of a profound transformation. This week, the Waltham, Mass.-based organization issued a number of key announcements, all focused on the same fundamental shift, as it readies the release of two commercial robots: Spot and Handle. The top of the company has recently seen its first major shakeup since its founding in the early 1990s. Founder Marc Raibert has stepped aside from his role as CEO, a transition that occurred quietly back in October. Longtime employee Rob Playter will be stepping into the position, having most recently served as Boston Dynamics’ COO. Playter joined the company early on, after penning a PhD thesis on “Passive Dynamics in the Control of Gymnastic Maneuvers.” A former NCAA gymnast himself, Playter’s work clearly has a bit of a spiritual successor in the complex maneuvers of the bipedal Atlas — arguably the most advanced of Boston Dynamics’ machines. The move comes during an important crossroads for the company, and Raibert, its longtime leader, public face and chief evangelist. “I just had my 70th birthday,” he tells TechCrunch. “So I’ve been thinking about this for about a year that we needed a succession plan and I had been working on it during that time, talking to SoftBank, making sure they were cool with the idea, and making sure I was cool with the idea.” Playter has hung with the company through numerous changes, serving as a Robotics Director at Google after the software giant purchased Boston Dynamics back in 2013. When the company switched hands to SoftBank, he took on the COO role. “I’ve been the organizational guy for a long time,” he says. “I basically hired most of the people in the company and growing us aggressively is a big challenge right now. Over the past year, bringing on new people into our executive leadership team has been a primary goal, as well as feeding an insatiable appetite for our technical teams to grow in order to meet the goals we’ve set for them. Which includes not only advancing the state of the art of robotics but actually making some of our robots into products and delivering them and supporting them and changing the organization to do so.” Focus for many at Boston Dynamics has shifted in recent years. At our Robotics event a few years back, Raibert announced plans to offer a commercial version of its Spot robot, aimed at performing security, construction site inspections and a variety of different tasks. Spot officially went on sale in September, and the company tells TechCrunch that it’s ramping up production with hopes of hitting 1,000 a year. Boston Dynamics has long insisted that the production version of Spot will be a platform, allowing end users to tailor it to a variety of different tasks. That dream takes a step closer to reality with the release of the Spot software development kit on GitHub this week. “The SDK enables a broad range of developers and non-traditional roboticists to communicate with the robot and develop custom applications that enable Spot to do useful tasks across a wide range of industries,” VP Michael Perry said in a statement offered to TechCrunch. “With the SDK, developers in the Early Adopter Program can create custom methods of controlling the robot, integrate sensor information into data analysis tools, and design custom payloads which expand the capabilities of the base robot platform.” The company has already announced a few early partners. Perry again, “One of our customers, HoloBuilder, is using the SDK to integrate Spot into their existing app. With what they’ve developed, workers can use a phone to teach Spot to document a path around a construction site and then Spot will autonomously navigate that path and take 360 images that go right into their processing software. Other customers are exploring VR control, automated registration of laser-scanning, connecting Spot’s data to cloud work order services, using Edge computing to help Spot semantically understand its environment, and much more.” In keeping with the company’s longstanding viral approach to marketing, buster of myths Adam Savage is among the early developers. Savage will participate in development with the robot for the next year, a milestone he celebrated with the release of a Tested video that understandably mostly entails him controlling the robot like a kid with a new RC car on Christmas morning. Ultimately, Boston Dynamics will only have so much input into what happens to these robots once they’re out in the world. There are currently nearly 100 robots out in the world, and production is ramping up to around 83 units a month. A video that debuted onstage at our robotics event last year recently caused a dust up with the ACLU after showcasing state police using Spot in hostage rescue field training. “There is a part of a humanity that loves to worry about robots taking over or being weaponized or something like that,” says Raibert. “We definitely want to counter that narrative. We’re not interested in weaponized robots. We’ve also gotten positive feedback from the fact that the police were using our robot to look at suspicious packages. There’s a real safety issue there and that it generated some additional interest with us as well. I mean, this isn’t really anything different than any new technology. There’s a wide variety of things it can be used for. We’re working to be responsible and trying find the good things that it could be used for.” SoftBank’s 2018 acquisition marked a major turning point for the company, of course. Though Boston Dynamics insists that the investor firm has done little in the way of pressing it into commercialization. Those plans, it explains, were already in the works. “I think the remarkable thing about SoftBank is they’re absolutely interested in the products and the moneymaking potential of what we’re doing while having a very serious interest in support for the longer range stuff we’re doing,” says Raibert, who is staying on at BD/SoftBank in a chairman role. “And over the 18 months that we’ve been part of SoftBank, I’ve repeatedly tested that commitment. Talking to the top leadership at SoftBank and they have repeatedly endorsed that. They’re very enthusiastic also for us to productize and make robots and make robot products.” The company will maintain its commitment to developing research robots, a role Raibert will continue to help facilitate. That list includes products like Atlas and, no doubt, some still unannounced products. Others, including Handle, will be developed with an eye toward production. In April, the company acquired Bay Area-based 3D vision startup Kinema Systems to bring imaging to the pick-and-place robot. Boston Dynamics tells TechCrunch that it plans to offer the robot up for commercial use in the next two years. “We’ve been doing proof of concept tests with Handle and some early customers to validate our expectations on what the useful tasks are in a warehouse for moving boxes around,” says Playter. “We have a small set of those customers and we’re getting feedback from them. So far they’re really excited about this capability. It’s unique. As far as I know, we’re the only case-picking warehouse robot in development right now. And this is just a ubiquitous job, whether you’re unloading trucks or loading trucks or building pallets or de-palletizing. There’s thousands of warehouses just full of boxes.”

Read More...
posted 5 days ago on techcrunch
Fintech startup Revolut lets you earn interests on your savings thanks to a new feature called savings vaults. That feature is currently only available to users living in the U.K. and paying taxes in the U.K. The company has partnered with Flagstone for that feature. For now, the feature is limited to Revolut customers with a Metal subscription (£12.99 per month or £116 per year). But Revolut says that it will be available to Revolut Premium and Standard customers in the near future. Savings vaults work pretty much like normal vaults. You can create sub-accounts in the Revolut app to put some money aside. And Revolut offers you multiple ways to save. You can round up all your card transactions to the nearest pound and save spare change in a vault. You can also set up weekly or monthly transactions from your main account to a vault. And of course, you can transfer money manually whenever you want. Metal customers in the U.K. can now turn normal vaults into savings vaults. The only difference is that you’re going to earn interests — Revolut pays those interests daily. You can take money from your savings vault whenever you want. Revolut promises 1.35% AER interest rate up to a certain limit. If you put a huge sum of money in your savings vault, you’ll get a lower interest rate above the limit. Your money is protected by the FSCS up to a value of £85,000 for eligible customers.

Read More...
posted 5 days ago on techcrunch
The food industry may be the biggest industry in the world, but it’s also one of the least efficient. BCG says 1.6 billions tons of food, worth $1.2 trillion, is wasted in food every year and those numbers are only expected to go up. A number of players have stepped up to try and solve their own portion of the problem, and one such solution is Crisp. The company, which received $14 million in Series A funding last year led by FirstMark Capital, is today going live with its platform (which has been in beta). Crisp aims to solve the global food waste problem via demand forecasts. Founder and CEO Are Traasdahl, a serial founder, believes that a lack of communication and data flow between the many players in the supply chain is a main cause for all this waste, a great deal of which happens long before the food reaches the consumer. Right now, forecasting demand is no where close to a perfect science for many of these players. From food brands to distributors to grocery stores, the problem is usually solved by looking at a spreadsheet from last year’s sales for hours to try to determine the signals that played into this or that SKU’s sales performance. And then there was Crisp. Integrated with almost any ERP software a company might have, Crisp ingests historical data from these food brands and combines that data with signals around other demand drivers, such as seasonality, holidays, price sensitivity and other pricing information, marketing campaigns, competitive landscape, weather that might affect the sale or shipment of certain produce or other ingredients. Using these data points, and historical sales data, Crisp believes it can give a much more accurate picture of demand over the next day, week, month or year. But Crisp isn’t just for food brands, such as Nounós Creamery, a Crisp customer that says its reduced scrapped inventory by 80 percent since switching to the platform. Crisp serves almost every player in the food supply chain, from retailers to distributors to brands to brokers. And the more customers it gets, the better it is at predicting demand on a very specific level. For instance, the demand forecasting Crisp offers for a particular grocery store, based on external data, will obviously get much better once that grocery store is a customer on the platform. Traasdahl was initially concerned that his customers would be reluctant to hand over this type of sensitive sales data, and also that players within the industry might be anxious to hand over such data to a platform that’s aggregating everyone’s data, including their competitors. Turns out, the food industry has more of a “better together” mentality. “Other industries are not as dependent on each other,” said Traasdahl. “If I am a creamery and need to buy blueberries for my yogurt, I may have five different vendors for those blueberries. And if they don’t get delivered on the right day, Costco will yell at me for being late with the yogurt. Everyone in the supply chain is somewhat dependent on each other.” For that reason, it’s been easier to attract clients to the platform than expected. The prospect of a collaborative demand forecast platform, that’s pulling signals from across the entire industry, is going to be more accurate than siloed demand forecasts produced by a single vendor or brand. During the beta program, which launched in October, Crisp brought on more than 30 companies to the platform, including Gilbert’s Craft Sausages, SunFed Perfect Produce, Nounós Creamery, Hofseth, REMA and Superior Farms.

Read More...
posted 5 days ago on techcrunch
Lighter Capital announced today that it has secured access to $100 million to lend to growing startups. The firm is best-known for its work with revenue-based financing, in which expanding companies repay borrowed funds out of future receipts. Lighter has also expanded into other, equity-free capital options for startups in the last year. Lighter is most easily understood as part of the group of firms that provide what TechCrunch has described as “alt-VC,” forms of capital access that do not fit into the traditional venture capital model of selling shares (equity) for cash. With the VC method, venture capitalists raise funds from wealthy capital pools, disbursing the funds in pieces to various private companies for an ownership stake. Those growth-focused firms then try to scale rapidly. Those that succeed become valuable, rendering the venture investment lucrative, and, hopefully, the venture capital fund profitable. In alt-VC, various forms of debt are put to work, tailored to companies that are growth-oriented, often existing outside of the realm of what traditional banks would consider lending-ready. Startups that are working in software-as-a-service (SaaS) or e-commerce are often considered ideal candidates for alt-VC in its various forms, as returns that can be generated with marginally deployed capital are calculable with reasonable certainty in those fields. Got all that? Let’s turn to what Lighter Capital is up to. Working capital Lighter’s new $100 million access to capital (we’ll call it a fund, for lack of a better term) will allow it to accelerate its business, the firm’s CEO Thor Culverhouse told TechCrunch. Lighter has a number of “ideas about how we’re going to grow [its] business,” Culverhouse said in a phone call, and having more “access to capital is a very important element to that growth strategy.” How some founders are raising capital outside of the VC world According to a release, Lighter has “invested” over $200 million in more than 350 companies to date; however, even though Lighter’s loans return capital and could allow for the recycling of funds, the $100 million in new funds represents a step up in capacity for the company. (Lighter is working with HCG for its capital access.) The new funds will be disbursed in more ways than one. In June of 2019, Lighter added two more traditional forms of debt to its list of offerings: term loans and lines of credit. Culverhouse discussed the additional products with TechCrunch, connecting term loans to revenue-based financing options: We did two things. When you think about the [revenue-based financing] function we have today, it is a term loan, it’s just that the repayment is based on whatever your monthly recurring revenue is. What we noticed is some people liked that flexibility. We [also] noticed some of our customers said, actually, I’d rather have a very predictable payment stream. And so we came out with another term loan that is like any other term loan, it’s just as a predictable payment stream throughout the year. So they’re very, very much alike. And then we came out with a line of credit, which is more traditionally used for working capital. So it’s a 12-month revolver, if you will. Here Lighter capital describes a link between revenue-based financing and regular loans that is worth chewing on. Revenue-based financing is merely a loan, tuned modestly for the SaaS world. That’s it. It allows for recurring-revenue focused companies to vary their payments over time, but both a term loan to a growth-oriented startup and a revenue-based financing event are pretty similar at their core. Which, naturally, makes Lighter’s move into more traditional loans pretty reasonable. With $100 million to put to work, Lighter is going to move some cash. That, in conjunction with the growing set of firms offering similar services, should help a lot of folks fund their companies’ growth without selling shares.

Read More...
posted 5 days ago on techcrunch
Twitter is pouring a little more fuel on the messaging fire. It’s added a heart+ button to its direct messaging interface which lets users shortcut to a pop-up menu of seven emoji reactions so they can quickly express how they’re feeling about a missive. Emoji reactions can be added to text or media messages — either via the heart+ button or by double tapping on the missive to bring up the reaction menu. The social network teased the incoming tweak a few hours earlier in a knowing tweet about sliding into DMs that actually revealed the full line-up of reaction emojis — which, in text form, can be described as: Crying lol; shocked/surprised; actually sad; heart; flame; thumb-up and thumb-down.   So instead of a smilie face Twitter users are being nudged towards an on-brand-message Twitter heart, in keeping with its long-standing pick for a pleasure symbol. The flame is perhaps slightly surprising for a company that’s publicly professed to wanting to improve the conversational health of its platform. If it’s there to stand in for appreciation a clap emoji could surely have done the trick. Whereas flame wars aren’t typically associated with constructive speech. But — hey — the flame icon does catch the eye… Say more with new emoji reactions for Direct Messages! To add a reaction, click the icon that appears when you hover over the message on web or double tap the message on mobile and select an emoji from the pop-up. For more about DM reactions: https://t.co/sdMumGDBYl https://t.co/QxMVmGt8eY — Twitter Support (@TwitterSupport) January 22, 2020 Twitter is late to this extroverted party. Rival messaging platforms such as Apple iMessage and Facebook Messenger have had emoji reactions for years, whereas Twitter kept things relatively minimal and chat-focused in its DM funnel — to its credit (at least if you value the service as, first and foremost, an information network). So some might say Twitter jumping on the emoji reaction bandwagon now is further evidence it’s trying to move closer to rivals like Facebook as a product. (See also: Last year’s major desktop product redesign — which has been compared in look and feel to the Facebook News Feed.) But if so this change is at least a relatively incremental one. Twitter users have also, of course, always been able to react to an incoming DM by sending whatever emoji or combination of emoji they prefer as a standard reply. Though now lazy thumbs have a shortcut to emote — so long as they’re down with Twitter’s choice of icons. In an FAQ about the new DM emoji reactions, Twitter notes that emoting will by default send a notification to all conversation participants “any time a new reaction is added to a message”. So, yes, there’s attention-spamming potential aplenty here… Adjust your notification and DM settings accordingly. You can only choose one reaction per missive. Each symbol is displayed under the message/media with a count next to it — to allow for group tallies to be totted up.  NB: Clicking on another symbol will swap out the earlier one — generating, er, more notification spam. And really annoying people could keep flipping their reaction to generate a real-time emoji streaming game of notification hell (hi growth hackers!) with folks they’ve been DMing with. So that’s another good reason to lock down your Twitter settings. Users still running older version of Twitter’s apps which don’t support message reactions will see a standard text emoji message per reaction sent (see examples below). This kinda confusingly makes it look like the reaction sender has actually been liking/flaming their own stuff. So all the more reason to not be spammy about emoji.

Read More...
posted 5 days ago on techcrunch
NASA has finalized the payloads for its first cargo deliveries scheduled to be carried by commercial lunar landers, vehicles created by companies the agency selected to take part in its Commercial Lunar Payload Services (CLPS) program. In total, there are 16 different payloads, which consist of a number of difference science experiments and technology experiments, that will be carried by landers built by Astrobotic and Intuitive Machines. Both of these landers are scheduled to launch next year, carrying their cargo to the Moon’s surface and helping prepare the way for NASA’s mission to return humans to the Moon by 2024. Astrobotic’s Peregrine is set to launch aboard a rocket provided by the United Launch Alliance (ULA), while Intuitive Machines’ Nova-C lander will make its own lunar trip aboard a SpaceX Falcon 9 rocket. Both landers will carry two of the payloads on the list, including a Laser Retro-Reflector Array (LRA) that is basically a mirror-based precision location device for situating the lander itself; and a Navigation Doppler Lidar for Precise Velocity and Range Sensing (NDL) – a laser-based sensor that can provide precision navigation during descent and touchdown. Both of these payloads are being developed by NASA to ensure safe, controlled and specifically targeted landing of spacecraft on the Moon’s surface, and their use here be crucial in building robust lunar landing systems to support Artemis through the return of human astronauts to the Moon and beyond. Besides those two payloads, everything else on either lander is unique to one vehicle or the other. Astrobotic is carrying more, but its Peregrine lander can hold more cargo – its payload capacity tops out at around 585 lbs, whereas the Nova-C can carry a maximum of 220 lbs. The full list of what each lander will have on board is available below, as detailed by NASA. Overall, NASA has 14 total contractors that could potentially provide lunar payload delivery services through its CLPS program. That basically amounts to a list of approved vendors, who then bid on whatever contracts the agency has available for this specific need. Other companies on the CLPS list include Blue Origin, Lockheed Martin, SpaceX and more. Starting with these two landers next year, NASA hopes to fly around two missions per year each year through the CLPS program. Astrobotic Payloads Surface Exosphere Alterations by Landers (SEAL): SEAL will investigate the chemical response of lunar regolith to the thermal, physical and chemical disturbances generated during a landing, and evaluate contaminants injected into the regolith by the landing itself. It will give scientists insight into the how a spacecraft landing might affect the composition of samples collected nearby. It is being developed at NASA Goddard. Photovoltaic Investigation on Lunar Surface (PILS): PILS is a technology demonstration that is based on an International Space Station test platform for validating solar cells that convert light to electricity. It will demonstrate advanced photovoltaic high-voltage use for lunar surface solar arrays useful for longer mission durations. It is being developed at Glenn Research Center in Cleveland. Linear Energy Transfer Spectrometer (LETS): The LETS radiation sensor will collect information about the lunar radiation environment and relies on flight-proven hardware that flew in space on the Orion spacecraft’s inaugural uncrewed flight in 2014. It is being developed at NASA Johnson. Near-Infrared Volatile Spectrometer System (NIRVSS): NIRVSS will measure surface and subsurface hydration, carbon dioxide and methane – all resources that could potentially be mined from the Moon — while also mapping surface temperature and changes at the landing site. It is being developed at Ames Research Center in Silicon Valley, California. Mass Spectrometer Observing Lunar Operations (MSolo): MSolo will identify low-molecular weight volatiles. It can be installed to either measure the lunar exosphere or the spacecraft outgassing and contamination. Data gathered from MSolo will help determine the composition and concentration of potentially accessible resources. It is being developed at Kennedy Space Center in Florida. PROSPECT Ion-Trap Mass Spectrometer (PITMS) for Lunar Surface Volatiles: PITMS will characterize the lunar exosphere after descent and landing and throughout the lunar day to understand the release and movement of volatiles. It was previously developed for ESA’s (European Space Agency) Rosetta mission and is being modified for this mission by NASA Goddard and ESA. Neutron Spectrometer System (NSS): NSS will search for indications of water-ice near the lunar surface by measuring how much hydrogen-bearing materials are at the landing site as well as determine the overall bulk composition of the regolith there. NSS is being developed at NASA Ames. Neutron Measurements at the Lunar Surface (NMLS): NMLS will use a neutron spectrometer to determine the amount of neutron radiation at the Moon’s surface, and also observe and detect the presence of water or other rare elements. The data will help inform scientists’ understanding of the radiation environment on the Moon. It’s based on an instrument that currently operates on the space station and is being developed at Marshall Space Flight Center in Huntsville, Alabama. Fluxgate Magnetometer (MAG): MAG will characterize certain magnetic fields to improve understanding of energy and particle pathways at the lunar surface. NASA Goddard is the lead development center for the MAG payload. Intuitive Machines Payloads Lunar Node 1 Navigation Demonstrator (LN-1): LN-1 is a CubeSat-sized experiment that will demonstrate autonomous navigation to support future surface and orbital operations. It has flown on the space station and is being developed at NASA Marshall. Stereo Cameras for Lunar Plume-Surface Studies (SCALPSS): SCALPSS will capture video and still image data of the lander’s plume as the plume starts to impact the lunar surface until after engine shut off, which is critical for future lunar and Mars vehicle designs. It is being developed at NASA Langley, and also leverages camera technology used on the Mars 2020 rover. Low-frequency Radio Observations for the Near Side Lunar Surface (ROLSES): ROLSES will use a low-frequency radio receiver system to determine photoelectron sheath density and scale height. These measurements will aide future exploration missions by demonstrating if there will be an effect on the antenna response or larger lunar radio observatories with antennas on the lunar surface. In addition, the ROLSES measurements will confirm how well a lunar surface-based radio observatory could observe and image solar radio bursts. It is being developed at NASA Goddard.

Read More...
posted 5 days ago on techcrunch
When the Department of Defense finally made a decision in October on the decade long, $10 billion JEDI cloud contract, it seemed that Microsoft had won. But nothing has been simple about this deal from the earliest days, so  it shouldn’t come as a surprise that last night Amazon filed a motion to stop work on the project until the court decides on its protest of the DoD’s decision. The company announced on November 22nd that it had filed suit in the U.S. Court of Federal Claims protesting the DoD’s decision to select Microsoft. Last night’s motion is an extension of that move to put the project on hold until the court decides on the merits of the case. Even after Microsoft wins, JEDI saga could drag on “It is common practice to stay contract performance while a protest is pending and it’s important that the numerous evaluation errors and blatant political interference that impacted the JEDI award decision be reviewed. AWS is absolutely committed to supporting the DoD’s modernization efforts and to an expeditious legal process that resolves this matter as quickly as possible,” a spokesperson said in a statement last night. As we previously reported, the statement echoes sentiments AWS CEO Andy Jassy made at a press event during AWS re:Invent in December: “I would say is that it’s fairly obvious that we feel pretty strongly that it was not adjudicated fairly,” he said. He added, “I think that we ended up with a situation where there was political interference. When you have a sitting president, who has shared openly his disdain for a company, and the leader of that company, it makes it really difficult for government agencies, including the DoD, to make objective decisions without fear of reprisal.” This is just the latest turn in a contract procurement process for the ages. It will now be up to the court to decide if the project should stop or not, and beyond that if the decision process was carried out fairly. Despite JEDI loss, AWS retains dominant market position

Read More...
posted 5 days ago on techcrunch
Wrist-worn fitness trackers tend to do a fine job when exercising outdoors. For those glued to gym equipment, however, things get trickier. GPS can’t really do its job detecting distances, and machines like the elliptic tend to be even trickier. Recent generations of the Apple Watch and WatchOS have worked to bridge the gap, with more sophisticated workout detection and the addition of GymKit in 2017. With the latter, Apple started working with equipment manufactures to ditch the 30-pin iPod machines for newer models that worked with Apple’s detection. This week, the company takes a step further by partnering with the gyms themselves. The new Apple Watch Connected program will launch with four partners. It’s a pretty diverse quartet, ranging from old-school to boutique, including Orangetheory, Basecamp, YMCA and Crunch Fitness. And like GymKit before it, there’s a pretty good chance it will take a while to make it to your neighborhood workout facility, unless you live in a handful of metro areas, including Manhattan and the Twin Cities. Apple GymKit is coming to a treadmill near you The program is designed to further bridge the gap between life inside and outside of the gym. It’s basically a four-legged stool, including GymKit-enabled equipment, an Apple Watch and iOS app (developed with Apple), accepting Apple Pay and, perhaps, most interesting of the bunch, “incentive programs.” How each chain opts to participate is up to the specifics of their own business model. Most notably, GymKit machines may not always be applicable, as in the case of Orangetheory, whose workouts are built around machine-shifting interval training. As such, the GymKit logging ultimately makes less sense. Getting back to the incentive program, that, too, will vary a bit, depending on the nature of the deal with the gym. Take Orangetheory. Here you can can basically use activity to earn stuff like Nike and Apple gift cards. In the case of Crunch, you can earn deductions from your fees, up to $300 over two years. At YMCA, earnings go to “community initiatives,” while Basecamp’s go back to paying off the value of the Apple Watch Series 5 GPS the gym provides. All in all, seems like a win-win for all parties. Apple gets more active engagement in a small but growing concentrated number of gyms, and gyms get to list an Apple partnership among their perks. GymKit partners, meanwhile, are set to sell a bunch more machines. There are somewhere between 50,000 and 100,000 GymKit machines currently out there, a list that doesn’t include recently announced partners like Woodway, Octane and TRUE Fitness.

Read More...
posted 5 days ago on techcrunch
TikTok, the fast-growing user-generated video app from China’s Bytedance, has been building a new music streaming service to compete against the likes of Spotify, Apple Music and Amazon Music. And today it’s announcing a deal that helps pave the way for a global launch of it. It has inked a licensing deal with Merlin, the global agency that represents tens of thousands of independent music labels and hundreds of thousands of artists, for music from those labels to be used legally on the TikTok platform anywhere that the app is available. The news is significant because this is the first major music licensing deal signed by TikTok as part of its wider efforts in the music industry. That includes both its mainstay short-form videos — where music plays a key role (the app, before it was acquired by Bytedance, was even called ‘Musically’) — as well as new music streaming services. Specifically, a source close to TikTok has confirmed to TechCrunch that this Merlin deal covers its upcoming music subscription service Resso. Resso was long rumoured and eventually spotted in the wild at the end of last year when Bytedance tested the app in India and Indonesia. Bytedance owns the Resso trademark, so it’s a good bet that it will make its way to more markets soon. (Possibly with features that differentiate this later entrant from others in the market? Recall Bytedance acquired an AI-based music startup called Jukedeck last year.) “Independent artists and labels are such a crucial part of music creation and consumption on TikTok,” said Ole Obermann, global head of music for Bytedance and TikTok, in a statement. “We’re excited to partner with Merlin to bring their family of labels to the TikTok community. The breadth and diversity of the catalogue presents our users with an even larger canvas from which to create, while giving independent artists the opportunity to connect with TikTok’s diverse community.” Music is a fundamental part of the TikTok experience, and this deal covers everything that’s there today — videos created by TikTok users, sponsored videos created for marketing — as well as whatever is coming up around the corner. A music streaming app, which TikTok has reportedly been gearing up to launch for some time, is one way that the company could help generate revenue. Despite being one of the most popular apps of 2019, monetisation has largely eluded the company up to now. One reason why monetising can’t happen is because of the lack of deals at the other end of the chain. As of December, TikTok had yet to sign any deals with the “majors” — Sony Music, Warner Music and Universal Music — and from what we understand Merlin is the first big deal of its kind of the company. However, there are signs that more such agreements may be coming soon. Obermann, who was hired away from Warner Music last year, in turn hired another former Warner colleague, Tracy Gardner, who now leads label licensing for the company. And just yesterday, the company opened an office in Los Angeles, the heart of the music industry. The move to bring more licensed music usage to TikTok (and other Bytedance apps) is significant for other reasons, too. On one hand, it’s about labels trying to evolve with the times, collecting revenues wherever audiences happen to be, whether that is in short-form user-generated video, in advertising that runs alongside that, or in a new music service capitalising on the new vogue for streamed media. “This partnership with TikTok is very significant for us,” said Jeremy Sirota, CEO, Merlin, in a statement. “We are seeing a new generation of music services and a new era of music-related consumption, much of it driven by the global demand for independent music. Merlin members are increasingly using TikTok for their marketing campaigns, and today’s partnership ensures that they and their artists can also build new and incremental revenue streams.” One the other hand, the deal is significant also because it underscores how TikTok is increasingly working to legitimise itself in the wider tech and media marketplace. While Bytedance’s acquisition of TikTok continues to face regulatory scrutiny, the company has been working on ways to assert its independence from China’s control, which has included many clarifications about where its content is hosted (not China! it says) and even a search for a new US-based CEO. On another front, more licensing deals should also help the company with the many legal and PR issues that have been hanging over it concerning how it pays out when music is used in its popular app.

Read More...
posted 5 days ago on techcrunch
Did you notice a recent change to how Google search results are displayed on the desktop? I noticed something last week — thinking there must be some kind of weird bug messing up the browser’s page rendering because suddenly everything looked similar: A homogenous sea of blue text links and favicons that, on such a large expanse of screen, come across as one block of background noise. I found myself clicking on an ad link — rather than the organic search result I was looking for. Here, for example, are the top two results for a Google search for flight search engine ‘Kayak’ — with just a tiny ‘Ad’ label to distinguish the click that will make Google money from the click that won’t… Turns out this is Google’s latest dark pattern: The adtech giant has made organic results even more closely resemble the ads it serves against keyword searches, as writer Craig Mod was quick to highlight in a tweet this week. There's something strange about the recent design change to google search results, favicons and extra header text: they all look like ads, which is perhaps the point? pic.twitter.com/TlIvegRct1 — Craig Mod (@craigmod) January 21, 2020 Last week, in its own breezy tweet, Google sought to spin the shift as quite the opposite — saying the “new look” presents “site domain names and brand icons prominently, along with a bolded ‘Ad’ label for ads”: Last year, our search results on mobile gained a new look. That’s now rolling out to desktop results this week, presenting site domain names and brand icons prominently, along with a bolded “Ad” label for ads. Here’s a mockup: pic.twitter.com/aM9UAbSKtv — Google SearchLiaison (@searchliaison) January 13, 2020 But Google’s explainer is almost a dark pattern in itself. If you read the text quickly you’d likely come away with the impression that it has made organic search results easier to spot since it’s claiming components of these results now appear more “prominently” in results. Yet, read it again, and Google is essentially admitting that a parallel emphasis is being placed — one which, when you actually look at the thing, has the effect of flattening the visual distinction between organic search results (which consumers are looking for) and ads (which Google monetizes). Another eagle-eyed user Twitter, going by the name Luca Masters, chipped into the discussion generated by Mod’s tweet — to point out that the tech giant is “finally coming at this from the other direction”. They're finally coming at this from the other direction:https://t.co/XYkHjVrE8X — Luca K. B. Masters (@lkbm) January 21, 2020 ‘This’ being deceptive changes to ad labelling; and ‘other direction’ being a reference to how now it’s organic search results being visually tweaked to shrink their difference vs ads. Google previously laid the groundwork for this latest visual trickery by spending earlier years amending the look of ads to bring them closer in line with the steadfast, cleaner appearance of genuine search results. Except now it’s fiddling with those too. Hence ‘other direction’. Masters helpfully quote-tweeted this vintage tweet (from 2016), by journalist Ginny Marvin — which presents a visual history of Google ad labelling in search results that’s aptly titled “color fade”; a reference to the gradual demise of the color-shaded box Google used to apply to clearly distinguish ads in search results. Those days are long gone now, though. Color fade: A history of Google ad labeling in search results https://t.co/guo3jc4kwz pic.twitter.com/LMYqhmgfyE — Ginny Marvin (@GinnyMarvin) July 25, 2016   Now a user of Google’s search engine has — essentially — only a favicon between them and an unintended ad click. Squint or you’ll click it. This visual trickery may be fractionally less confusing in a small screen mobile environment — where Google debuted the change last year. But on a desktop screen these favicons are truly minuscule. And where to click to get actual information starts to feel like a total lottery. A lottery that’s being stacked in Google’s favor because confused users are likely to end up clicking more ad links than they otherwise would, meaning it cashes in at the expense of web users’ time and energy. Back in May, when Google pushed this change on mobile users, it touted the tweaks as a way for sites to showcase their own branding, instead of looking like every other blue link on a search result page. But it did so while simultaneously erasing a box-out that it had previously displayed around the label ‘Ad’ to make it stand out. That made it “harder to differentiate ads and search results,” as we wrote then — predicting it will “likely lead to outcry”. There were certainly complaints. And there will likely be more now — given the visual flattening of the gap between ad clicks and organic links looks even more confusing for users of Google search on desktop. (Albeit, the slow drip of design change updates also works against mass user outcry.) Google’s new look for mobile search results puts site owners and publishers first We reached out to Google to ask for a response to the latest criticism that the new design for search results makes it almost impossible to distinguish between organic results and ads. But the company ignored repeat requests for comment. Of course it’s true that plenty of UX design changes face backlash, especially early on. Change in the digital realm is rarely instantly popular. It’s usually more ‘slow burn’ acceptance. But there’s no consumer-friendly logic to this one. (And the slow burn going on here involves the user being cast in the role of the metaphorical frog.) Instead, Google is just making it harder for web users to click on the page they’re actually looking for — because, from a revenue-generating perspective, it prefers them to click an ad. It’s the visual equivalent of a supermarket putting a similarly packaged own-brand right next to some fancy branded shampoo on the shelf — in the hopes a rushed shopper will pluck the wrong one. (Real life dark patterns are indeed a thing.) It’s also a handy illustration of quite how far away from the user Google’s priorities have shifted, and continue to drift. “When Google introduced ads, they were clearly marked with a label and a brightly tinted box,” said UX specialist Harry Brignull. “This was in stark contrast to all the other search engines at the time, who were trying to blend paid listings in amongst the organic ones, in an effort to drive clicks and revenue. In those days, Google came across as the most honest search engine on the planet.” Brignull is well qualified to comment on dark patterns — having been calling out deceptive design since 2010 when he founded darkpatterns.org. “I first learned about Google in the late 1990s. In those days you learned about the web by reading print magazines, which is charmingly quaint to look back on. I picked up a copy of Wired Magazine and there it was – a sidebar talking about a new search engine called ‘Google’,” he recalled. “Google was amazing. In an era of portals, flash banners and link directories, it went in the opposite direction. It didn’t care about the daft games the other search engines were playing. It didn’t even seem to acknowledge they existed. It didn’t even seem to want to be a business. It was a feat of engineering, and it felt like a public utility. “The original Google homepage was recognised a guiding light of purism in digital design. Search was provided by an unstyled text field and button. There was nothing else on the homepage. Just the logo. Search results were near-instant and they were just a page of links and summaries – perfection with nothing to add or take away. The back-propagation algorithm they introduced had never been used to index the web before, and it instantly left the competition in the dust. It was proof that engineers could disrupt the rules of the web without needing any suit-wearing executives. Strip out all the crap. Do one thing and do it well.” “As Google’s ambitions changed, the tinted box started to fade. It’s completely gone now,” Brignull added. The one thing Google very clearly wants to do well now is serve more ads. It’s chosen to do that deceptively, by steadily — and consistently — degrading the user experience. So a far cry from “public utility”. And that user-friendly Google of old? Yep, also completely gone.

Read More...
posted 5 days ago on techcrunch
When Dfinity raised $102 million in funding in 2018 at a $2 billion valuation in a round jointly led by Andreessen Horowitz and Polychain Capital, it was thought of as a step change in the world of blockchain technology. In an area that was  synonymous generating a lot of headlines around cryptocurrency speculation, this was a shift in focus, looking instead at the architecture behind Bitcoin, Ethereum, and the rest, and how it could be used for more than just “mining”, distributing and using new financial instruments — with a major, mainstream VC backing the idea, no less. Dfinity launched with a very lofty goal: to build what it called the “Internet Computer”: a decentralized and non-proprietary network to run the next generation of mega-applications. It dubbed this public network “Cloud 3.0”. Now, looks like this is Cloud is now about to break. In Davos this week, Dfinity launched the Bronze edition of its Internet Computer, a limited release that takes the startup one step closer to its full commercial release, expected later this year. And to prove out the concept of how an application would run on its new network, Dfinity today demonstrated an open social network called LinkedUp. The start-up has rather cheekily called this “an open version of LinkedIn,” the Microsoft-owned social network for professional. Unlike LikedIn, LinkedUp — which runs on any browser, is not owned or controlled by a corporate entity. LinkedUp is built on Dfinity’s co-called Internet Computer, its name for the platform it is building to distribute the next generation of software and open internet services. The software is hosted directly on the internet on a Switzerland-based independent data center, but in the concept of the Internet Computer, it could be hosted at your house or mine: the compute power to run the application — LinkedUp, in this case — is coming not from Amazon AWS, Google Cloud or Microsoft Azure, and is instead based on the distributed architecture that Dfinity is building. Dfinity is open-sourcing LinkedUp for developers to create other types of open internet services on the structure it has built. This ‘open social network for professional profiles’ suggests that, on Difinity’s opensource software, one could create an ‘Open WhatsApp’, ‘Open eBay’, ‘Open Salesforce’, or ‘Open Facebook’. (Good news, since LinkedIn might not be so happy about a lookalike service with a name and layout that also looks very familiar. “While we can’t comment specifically on any proposed trademark, LinkedIn does monitor and take action as necessary to protect our trademarks,” a spokesperson said.) “Big tech has hijacked the internet and stifled innovation by owning the proprietary infrastructure and user relationships,” said Dominic Williams, Founder and Chief Scientist at Dfinity in a statement. “As a result, a handful of for-profit companies have created a monopolistic and closed internet. The Internet Computer provides a means to rebuild internet services in open form.” So perhaps what we should be calling this is not LinkedUp, but more a new sort of “Linux for the cloud”. Dfinity claims the application was built by “1.5 engineers in three weeks,” thus demonstrating how easy the infrastructure is to use. The tools include a Canister Software Developer Kit and a simple programming language called Motoko that is optimized for Dfinity’s Internet Computer. “The Internet Computer is conceived as an alternative to the $3.8 trillion dollar legacy IT stack, and empower the next-generation of developers to build a new breed of tamper-proof enterprise software systems and open internet services. We are democratizing software development,” Williams said. “The Bronze release of the Internet Computer provides developers and enterprises a glimpse into the infinite possibilities of building on the Internet Computer — which also reflects the strength of the Dfinity team we have built so far.” Dfinity says its “Internet Computer Protocol” allows for a new type of software called autonomous software, which can guarantee permanent APIs that cannot be revoked. When all these open internet services (e.g., open versions of WhatsApp, Facebook, eBay, Salesforce, etc) are combined with other open software and services it creates “mutual network effects” where everyone benefits. We quizzed Dfinity a little more on all this and asked whether this was an actual launch. A spokesperson told us: “Since our first major milestone of launching a terminal-based SDK and new programming language called Motoko — by the co-creator of WebAssembly — on 1 November, DFINITY has released 13 new public versions of the SDK, to our second major milestone [at WEF Davos] of demoing a decentralized web app called LinkedUp on the Internet Computer running on an independent data center in Switzerland. Subsequent milestones towards the public launch of the Internet Computer will involve (1) on-boarding a global network of independent data centers, (2) fully tested economic system, and (3) fully tested Network Nervous Systems for configuration and upgrades.” It also looks like Dfinity will not be raising more money just yet. But the question is how they plan to woo people to it? “Dfinity has been working with a select group of Fortune 500 companies, strategic consultancies, systems integrators, venture capitalists, and universities,” the company said. We are not sure that will quite suffice to take out Facebook, LinkedIn and all the other tech giants, but we’re fascinated to see how this plays out.

Read More...
posted 6 days ago on techcrunch
Language-learning platform Busuu, which has fast expanded to take on traditional giants like Duolingo, says it has acquired the live video tutoring company Verbling for an undisclosed amount, other than calling it a “double-digit million dollar acquisition.” As a result, Busuu will now use the Verbling platform to expand into the live video tutoring space for its consumer users and corporate clients. Busuu says it recently surpassed 100 million users globally, makes it one of the world’s fastest-growing EdTech companies. It says it reach cash flow break-even last year, and plans to generate over $40 million in revenues in 2020. CEO and cofounder Bernhard Niesner said “we also plan to go public in the future.” Speaking to TechCrunch, he said: “We are operating in the massive $60bn global language learning market, with digital language learning only representing a tiny 10% market share right now. This digital part will grow fast due to wider consumer adoption driven by better learning outcomes, expected to reach $17bn market value in 2027. Getting access to the capital markets would allow us to accelerate our growth, expand into other learning areas and build a truly globally leading, multi-billion dollar, digital learning business.” The new Verbling-based ‘Busuu live’ will be a combination of their AI-powered learning content, interaction with other learners plus 1-1 live tutoring with professional teachers. “We are also excited to leverage our 4bn data points from our learners to provide useful information to our new 10,000+ live teachers about their students. So whenever a teacher starts a live lesson, they will have access to relevant information about the progress of their students within Busuu, so they can fully adapt their lessons to the individual needs of their learners.” Busuu was originally founded in Madrid in 2008 and in 2012 moved to London, but now plans to open an office back in its ‘home town.’ Niesner said: “The London hiring market has become increasingly more competitive over the last couple of years (also due to Brexit, competition from Facebook and Google etc) while the Spanish startup-ecosystem has made tremendous progress.” Verbling was founded in San Francisco in 2011 by the Swedish co-founders Mikael Bernstein (CEO) and Gustav Rydstedt (CTO) who met while studying at Stanford University. After attending the Y-Combinator program, Verbling raised over $4.4m from Learn Capital, DFJ and Bullpen Capital. The platform has over 10,000 pre-vetted live teachers and offers interactive 1-1 lessons in nearly 60 different languages. Mikael Bernstein, Co-Founder and CEO, Verbling said: “We are very excited to be joining forces with Busuu’s talented and experienced team, combining our world-class tutors with Busuu’s AI-powered platform will enable language learners across the globe to reach proficiency even faster.” Following the acquisition, Verbling’s team members, including co-founders Mikael Bernstein (CEO) and Gustav Rydstedt (CTO) will join Busuu. For context, the main publicly-listed language learning business is Rosetta Stone but they belong to the old version of language learning and have not yet done their shift to mobile, although they might survive that. There are expectations that both Duolingo and VIPKids (the Chinese English learning unicorn) will go public soon.

Read More...
posted 6 days ago on techcrunch
Internet services company Opera has come under a short-sell assault based on allegations of predatory lending practices by its fintech products in Africa. Hindenburg Research issued a report claiming (among other things) that Opera’s finance products in Nigeria and Kenya have run afoul of prudent consumer practices and Google Play Store rules for lending apps. Hindenburg — which is based in NYC and managed by financial analyst Nate Anderson — went on to suggest Opera’s U.S. listed stock was grossly overvalued. That’s a primer on the key info, though there are several additional shades of the who, why, and where of this story to break down, before getting to what Opera and Hindenburg had to say. A good start is Opera’s ownership and scope. Founded in Norway, the company is an internet services provider, largely centered around its Opera browser. Opera was acquired in 2016 for $600 million by a consortium of Chinese investors, led by current Opera CEO Yahui Zhou. Two years later, Opera went public in an IPO on NASDAQ, where its shares currently trade. Though Opera’s web platform isn’t widely used in the U.S. — where it has less than 1% of the browser market — it has been number-one in Africa, and more recently a distant second to Chrome, according to StatCounter. On the back of its browser popularity, Opera went on an African venture-spree in 2019, introducing a suite of products and startup verticals in Nigeria and Kenya, with intent to scale more broadly across the continent. In Nigeria these include motorcycle ride-hail service ORide and delivery app OFood. Central to these services are Opera’s fintech apps: OPay in Nigeria and OKash and Opesa in Kenya — which offer payment and lending options. Fintech focused VC and startups have been at the center of a decade long tech-boom in several core economies in Africa, namely Kenya and Nigeria. In 2019 Opera led a wave of Chinese VC in African fintech, including $170 million in two rounds to its OPay payments service in Nigeria. Opera’s Africa fintech startup OPay gains $120M from Chinese investors Opera’s fintech products in Africa (as well as Opera’s Cashbean in India) are at the core of Hindenburg Research’s brief and short-sell position.  The crux of the Hindenburg report is that due to the declining market-share of its browser business, Opera has pivoted to products generating revenue from predatory short-term loans in Africa and India at interest rates of 365 to 876%, so Hindenburg claims. The firm’s reporting goes on to claim Opera’s payment products in Nigeria and Kenya are afoul of Google rules. “Opera’s short-term loan business appears to be…in violation of the Google Play Store’s policies on short-term and misleading lending apps…we think this entire line of business is at risk of…being severely curtailed when Google notices and ultimately takes corrective action,” the report says. Based on this, Hindenburg suggested Opera’s stock should trade at around $2.50, around a 70% discount to Opera’s $9 share-price before the report was released on January 16. Hindenburg also disclosed the firm would short Opera. Founder Nate Anderson confirmed to TechCrunch Hindenburg continues to hold short positions in Opera’s stock — which means the firm could benefit financially from declines in Opera’s share value. The company’s stock dropped some 18% the day the report was published. On motivations for the brief, “Technology has catalyzed numerous positive changes in Africa, but we do not think this is one of them,” he said. “This report identified issues relating to one company, but what we think will soon become apparent is that in the absence of effective local regulation, predatory lending is becoming pervasive across Africa and Asia…proliferated via mobile apps,” Anderson added. While the bulk of Hindenburg’s critique was centered on Opera, Anderson also took aim at Google. “Google has become the primary facilitator of these predatory lending apps by virtue of Android’s dominance in these markets. Ultimately, our hope is that Google steps up and addresses the bigger issue here,” he said. TechCrunch has an open inquiry into Google on the matter. In the meantime, Opera’s apps in Nigeria and Kenya are still available on GooglePlay, according to Opera and a cursory browse of the site. For its part, Opera issued a rebuttal to Hindenburg and offered some input to TechCrunch through a spokesperson. In a company statement opera said, “We have carefully reviewed the report published by the short seller and the accusations it put forward, and our conclusion is very clear: the report contains unsubstantiated statements, numerous errors, and misleading conclusions regarding our business and events related to Opera.” Opera added it had proper banking licenses in Kenyan or Nigeria. “We believe we are in compliance with all local regulations,” said a spokesperson. TechCrunch asked Hindenburg’s Nate Anderson if the firm had contacted local regulators related to its allegations. “We reached out to the Kenyan DCI three times before publication and have not heard back,” he said. As it pertains to Africa’s startup scene, there’ll be several things to follow surrounding the Opera, Hindenburg affair. The first is how it may impact Opera’s business moves in Africa. The company is engaged in competition with other startups across payments, ride-hail, and several other verticals in Nigeria and Kenya. Being accused of predatory lending, depending on where things go (or don’t) with the Hindenburg allegations, could put a dent in brand-equity. There’s also the open question of if/how Google and regulators in Kenya and Nigeria could respond. Contrary to some perceptions, fintech regulation isn’t non-existent in both countries, neither are regulators totally ineffective. Kenya passed a new data-privacy law in November and Nigeria recently established guidelines for mobile-money banking licenses in the country, after a lengthy Central Bank review of best digital finance practices. Nigerian regulators demonstrated they are no pushovers with foreign entities, when they slapped a $3.9 billion fine on MTN over a regulatory breach in 2015 and threatened to eject the South African mobile-operator from the country. As for short-sellers in African tech, they are a relatively new thing, largely because there are so few startups that have gone on to IPO. In 2019, Citron Research head and activist short-seller Andrew Left — notable for shorting Lyft and Tesla — took short positions in African e-commerce company Jumia, after dropping a report accusing the company of securities fraud. Jumia’s share-price plummeted over 50% and has only recently begun to recover. As of Wednesday, there were signs Opera may be shaking off Hindenburg’s report — at least in the market — as the company’s shares had rebounded to $7.35. Revisiting Jumia’s JForce scandal and Citron’s short-sell claims

Read More...
posted 6 days ago on techcrunch
Bounce, a Bangalore-based startup that operates over 15,000 electric and gasoline docked bikes and scooters in nearly three dozen cities in India, said today it has raised $105 million in a new funding round as it explores sustainable ways to expand within the nation and build its own electric vehicles. The new financing round, Series D, was co-led by existing investors Eduardo Saverin’s B Capital and Accel Partners India, said the startup. The new round valued Bounce at a little over $500 million, up from about $200 million in June last year, a person familiar with the matter told TechCrunch. TechCrunch reported in late November that Bounce was in advanced stages of talks to raise $150 million at over $500 million valuation. The new round pushes the startup’s total raise to $194 million. Bounce, formerly known as Metro Bikes, allows customers to rent a scooter and pay 10.5 Indian rupees (15 cents) for each kilometer of the ride. The platform clocks 1.2 million rides each day. Bounce earlier deployed its own operations team in each city and flooded the market with its scooters, but in recent weeks it has started to explore a new strategy, said co-founder and chief executive Vivekananda Hallekere in an interview with TechCrunch. “We realized that it was not the smartest move to expand Bounce’s network on our own,” he said. The startup now works with mom-and-pop stores and local merchants in each city and they run their own operations. To date, Bounce has replicated this model in six cities in India (including Vijayawada and Mangalore) and has partnered with over 250,000 shops and merchants. “We launch in the cities with our own vehicles, but overtime, these micro-entrepreneurs deploy their own bikes and scooters. They are still using our app, and are part of the Bounce platform, but they don’t have to be locked into our scooter ecosystem,” he explained. The shift in strategy comes as Bounce looks to cut expenses and find a sustainable way to expand. “Otherwise, I would need a billion dollar of debt to launch a million vehicles in India,” he argued. “We wanted a model that is scalable and profitable, and helps us create the most impact.” Bounce is part of a small group of startups that is attempting to address a market that cab-hailing services Uber and Ola have been unable to address. The startup competes with Vogo, which is backed by Ola, and Yulu, which maintains a partnership with Uber. Riding these bikes is more affordable than hailing a cab, and also two wheels are much faster in crowded traffic of urban cities than four. These bikes have also proven useful in other ways. Hallekere said female users account for over 30% of all rides on Bounce — a figure that beats the industry estimates, because women feel much more safer with bikes, he said. “They don’t have to worry about how they would commute back from work,” he said. Bounce is also working on building its own ecosystem of electric vehicles. The startup said it has already built a scooter with metal chassis that can survive for at least 200,000 kilometers. The idea is to build electric scooters that work best for shared mobility, something Hallekere said the ecosystem is currently missing. “In our tests, we found that even if you threw this bike from the first floor of a building, nothing happens to it. It is also more tech-enabled, so it can tell when the second seat of the bike is in use and can bill users accordingly,” he explained. The startup plans to deploy these vehicles in the coming months. Kabir Narang, a partner at B Capital, told TechCrunch in an interview that he sees great potential in the shared mobility future in India, and Bounce team’s passion and commitment to solving these challenges made it easy for them to place “long-term” bet on the startup.

Read More...
posted 6 days ago on techcrunch
If you’ve ever entered a company’s office as a visitor or contractor, you probably know the routine: check in with a receptionist, figure out who invited you, print out a badge and get on your merry way. Brussels, Belgium- and New York-based Proxyclick aims to streamline this process, while also helping businesses keep their people and assets secure. As the company announced today, it has raised a $15 million Series B round led by Five Elms Capital, together with previous investor Join Capital. In total, Proxyclick says it’s systems have now been used to register over 30 million visitors in 7,000 locations around the world. In the UK alone, over 1,000 locations use the company’s tools. Current customers include L’Oreal, Vodafone, Revolut, PepsiCo and Airbnb, as well as a number of other Fortune 500 firms. Gregory Blondeau, founder and CEO of Proxyclick, stresses that the company believes that paper logbooks, which are still in use in many companies, are simply not an acceptable solution anymore, not in the least because that record is often permanent and visible to other visitors. Proxyclick’s founding team. “We all agree it is not acceptable to have those paper logbooks at the entrance where everyone can see previous visitors,” he said. “It is also not normal for companies to store visitors’ digital data indefinitely. We already propose automatic data deletion in order to respect visitor privacy. In a few weeks, we’ll enable companies to delete sensitive data such as visitor photos sooner than other data. Security should not be an excuse to exploit or hold visitor data longer than required.” What also makes Proxyclick stand out from similar solutions is that it integrates with a lot of existing systems for access control (including C-Cure and Lenel systems). With that, users can ensure that a visitor only has access to specific parts of a building, too. In addition, though, it also supports existing meeting rooms, calendaring and parking systems and integrates with Wi-Fi credentialing tools so your visitors don’t have to keep asking for the password to get online. Like similar systems, Proxyclick provides businesses with a tablet-based sign-in service that also allows them to get consent and NDA signatures right during the sign-in process. If necessary, the system can also compare the photos it takes to print out badges with those on a government-issued ID to ensure your visitors are who they say they are. Blondeau noted that the whole industry is changing, too. “Visitor management is becoming mainstream, it is transitioning from a local, office-related subject handled by facility managers to a global, security and privacy driven priority handled by Chief Information Security Officers. Scope, decision drivers and key people involved are not the same as in the early days,” he said. It’s no surprise then that the company plans to use the new funding to accelerate its roadmap. Specifically, it’s looking to integrate its solution with more third-party systems with a focus on physical security features and facial recognition, as well as additional new enterprise features.

Read More...
posted 6 days ago on techcrunch
If I watch a Story cross-posted from Instagram to Facebook on either of the apps, it should appear as “watched” at the back of the Stories row on the other app. Why waste my time showing me Stories I already saw? It’s been over two years since Instagram Stories launched cross-posting to Stories. Countless hours of each feature’s 500 million daily users have been squandered viewing repeats. Facebook and Messenger already synchronized the watched/unwatched state of Stories. It’s long past time that this was expanded to encompass Instagram. I asked Facebook and Instagram if it had plans for this. A company spokesperson told me that it built cross-posting to make sharing easier to people’s different audiences on Facebook and Instagram, and it’s continuing to explore ways to simplify and improve Stories. But they gave no indication that Facebook realizes how annoying this is or that a solution is in the works. The end result if this gets fixed? Users would spend more time watching new content, more creators would feel seen, and Facebook’s choice to jam Stories in all its apps would fee less redundant and invasive. If I send a reply to a Story on one app, I’m not going to send it or something different when I see the same Story on the other app a few minutes or hours later. Repeated content leads to more passive viewing and less interactive communication with friends, despite Facebook and Instagram stressing that its this zombie consumption that’s unhealthy. The only possible downside to changing this could be fewer Stories ad impressions if secondary viewings of peoples’ best friends’ Stories keep them watching more than new content. But prioritizing making money over the user experience is again what Mark Zuckerberg has emphasized is not Facebook’s strategy. There’s no need to belabor the point any further. Give us back our time. Stop the reruns.

Read More...
posted 6 days ago on techcrunch
After announcing a $550 million fundraise last August, U.K. AI-based health services startup Babylon Health is putting some of that money to use with its widest-ranging project to date. The company has inked a 10-year deal with the city of Wolverhampton in England to provide an integrated health app covering 300,000 people, the entire population of the city. The financial terms of the deal are not being disclosed, but Babylon confirmed that the NHS is not taking a stake in the startup as part of it. The plan is to start rolling out the first phase of the app by the end of this year. Babylon Health is known for building AI-based platforms that help diagnose patients’ issues. Babylon’s services are provided as a complement to seeing actual clinicians — the idea being that the interactions and AI can speed up some of the work of getting people seen and into the system. Some of Babylon’s best known work to date has been a chatbot that it built for the NHS in the U.K., and, in addition to working with a number of private businesses on their employee healthcare services, it is also now in the process of rolling out services in 11 countries in Asia. (In August, Babylon said it was delivering 4,000 clinical consultations each day, or one patient interaction every 10 seconds; covering 4.3 million people worldwide; with more than 1.2 million digital consultations completed to date.) Even with all these milestones passed — milestones that have helped catapult Babylon to a $2 billion valuation — its latest project will be its most ambitious to date: it will be the first time that Babylon works on a project that combines both hospital and primary medical care into an all-in-one app. “We are extremely proud of this exciting 10-year partnership with RWT which will benefit patients and the NHS as a whole,” said Ali Parsa, CEO and founder of Babylon, in a statement. “We have over 1,000 AI experts, clinicians, engineers and scientists who will be helping to make Digital-First Integrated Care a reality and provide fast, effective, proactive care to patients. Together with RWT, we can demonstrate this works and help the NHS lead healthcare across the world.” The plan is for Babylon and the Royal Wolverhampton NHS Trust — the local health authority and body that will oversee the work for the city’s population — to build an app that will not only provide remote diagnoses, but also live monitoring of patients with chronic conditions (using wearables and other monitoring apps) and the ability to connect people with doctors and others remotely. Other services will include the ability to let patients access their own medical records and review their own consultations; book appointments; renew prescriptions; view a “digital twin” of their own state of health based on medical history and other details; and manage their rehab after a procedure, illness or injury. The gap in the market that Babylon is tackling is the fact that many countries are seeing populations that are both growing bigger and generally living longer, and that is putting a strain not just on public health services, but also those that are managed completely or partly privately. This has been a particularly painful theme in Babylon’s home market, the U.K., where healthcare is nationalised and is regularly facing budgetary and human capital shortages, but there is no infrastructure (or consumer finance) to supplement that for the majority of people. The aim, however, goes beyond simply filling NHS gaps; it’s also about trying to build services that fit better with how people live, for example to provide them with certain services at home to save them from coming into, say, a hospital to be treated if the condition merits it. “We know from our active engagement with patients of all ages and backgrounds that they are keen to use technology that will improve access and give them greater control of their own health, wellbeing and social inclusion,” said Trust Chief Executive David Loughton, CBE, in a statement. “For example, it should be normal for a patient with a long-term condition to take a blood-test at home, have the results fed into their app which alerts the specialist if they need an appointment. The patient chooses a time to meet, has the consultation through the app, works with their specialist to build a care plan, and the app encourages them to complete it whilst assessing the impact it’s having. This is our vision for properly joined-up and integrated care.” AI has become a major theme in the drive to improve healthcare and medicine overall, primarily through two main areas: providing diagnostic and other services to patients in situations, acting in roles that would otherwise be played by humans; and in research, acting as a “super brain” to help perform complex calculations in the quest for better drug discovery, disease pathology and other areas that would take humans far longer to do on their own. Well aware of the strains on health systems, startups, investors and other stakeholders have jumped into using AI in the hopes of creating more efficiency and potentially better outcomes. But that doesn’t mean that all the outcomes have actually been better. Google’s DeepMind encountered a lot of controversy around how it handled patient data in its own NHS deals, leading to questions and investigations that have now stretched into years. And BenevolentAI — which has been working on drug discovery — found itself raising money last year in round that devalued the loss-making company by half. Paul Bate, Babylon’s MD of NHS services, noted in an interview that Babylon is mindful of patient privacy and consent, and notes that the service is opt-in and transparent in its data usage when engaging users. He declined to comment on how and when data will be retained by the NHS or by Babylon (or both) but said it would be made clear in the app when it is launched. “It’s not a simple answer to say whether one body or another will keep it, but it will be transparent, both for US and the NHS, when it launches,” he added.

Read More...
posted 6 days ago on techcrunch
If robots are to help out in places like hospitals and phone repair shops, they’re going to need a light touch. And what’s lighter than not touching at all? Researchers have created a gripper that uses ultrasonics to suspend an object in midair, potentially making it suitable for the most delicate tasks. It’s done with an array of tiny speakers that emit sound at very carefully controlled frequencies and volumes. These produce a sort of standing pressure wave that can hold an object up or, if the pressure is coming from multiple directions, hold it in place or move it around. This kind of “acoustic levitation,” as it’s called, is not exactly new — we see it being used as a trick here and there, but so far there have been no obvious practical applications. Marcel Schuck and his team at ETH Zürich, however, show that a portable such device could easily find a place in processes where tiny objects must be very lightly held. A small electric component, or a tiny oiled gear or bearing for a watch or micro-robot, for instance, would ideally be held without physical contact, since that contact could impart static or dirt to it. So even when robotic grippers are up to the task, they must be kept clean or isolated. Acoustic manipulation, however, would have significantly less possibility of contamination. Another, more sinister-looking prototype. The problem is that it isn’t obvious exactly which combination of frequencies and amplitudes are necessary to suspend a given object in the air. So a large part of this work was developing software that can easily be configured to work with a new object, or programmed to move it in a specific way — rotating, flipping or otherwise moving it at the user’s behest. A working prototype is complete, but Schuck plans to poll various industries to see whether and how such a device could be useful to them. Watchmaking is of course important in Switzerland, and the parts are both small and sensitive to touch. “Toothed gearwheels, for example, are first coated with lubricant, and then the thickness of this lubricant layer is measured. Even the faintest touch could damage the thin film of lubricant,” he points out in the ETHZ news release. How would a watchmaker use such a robotic arm? How would a designer of microscopic robots, or a biochemist? The potential is clear, but not necessarily obvious. Fortunately, he has a bit of fellowship cash to spend on the question and hopes to spin it off as a startup next year if his early inquiries bear fruit. Softly, softly, catchy jelly: This ‘ultragentle’ robotic gripper collects fragile marine life

Read More...
posted 6 days ago on techcrunch
Memphis Meats, a developer of technologies to manufacture meat, seafood and poultry from animal cells, has raised $161 million in financing from investors including Softbank Group, Norwest and Temasek, the investment fund backed by the government of Singapore. The investment brings the company’s total financing to $180 million. Previous investors include individual and institutional investors like Richard Branson, Bill Gates, Threshold Ventures, Cargill, Tyson Foods, Finistere, Future Ventures, Kimbal Musk, Fifty Years and CPT Capital. Other companies including Future Meat Technologies, Aleph Farms, Higher Steaks, Mosa Meat and Meatable are pursuing meat grown from cell cultures as a replacement for animal husbandry, whose environmental impact is a large contributor to deforestation and climate change around the world. Lab-grown meat could be on store shelves by 2022, thanks to Future Meat Technologies Innovations in computational biology, bio-engineering and materials science are creating new opportunities for companies to develop and commercialize technologies that could replace traditional farming with new ways to produce foods that have a much lower carbon footprint and bring about an age of superabundance, according to investors. What’s beyond Beyond Meat and Impossible Foods in the future of food? The race is on to see who will be the first to market with a product. “For the entire industry, an investment of this size strengthens confidence that this technology is here today rather than some far-off future endeavor. Once there is a “proof of concept” for cultivated meat — a commercially available product at a reasonable price point — this should accelerate interest and investment in the industry,” said Bruce Friedrich, the executive director of the Good Food Institute, in an email. “This is still an industry that has sprung up almost overnight and it’s important to keep a sense of perspective here. While the idea of cultivated meat has been percolating for close to a century, the very first prototype was only produced six years ago.”

Read More...
posted 6 days ago on techcrunch
Back in September, Betaworks put out a call for startups to participate in its latest “camp,” this one focused on audio. Danika Laszuk, the head of Betaworks Camp, told me at the time that the startup studio was looking for companies that are trying to build “audio-first” experiences for smart speakers and wireless headphones, or pursuing other audio-related opportunities like synthetic audio or social audio. Now Betaworks is unveiling the six startups that it has selected to participate in the program, covering everything from game assistants to AI music production. Each startup receives a pre-seed investment from Betaworks, and will be working out of the firm’s New York City offices for the next three months. Here are the companies: Storm is working on a live audio platform that it says will allow your friends to ask you anything. Midgame is building voice-enabled gaming assistants, starting with a bot that answers questions to improve your gameplay in Stardew Valley. Scout FM is developing hands-free listening experiences such as podcast radio stations and voice assistants for Amazon Alexa, Apple CarPlay and Android Auto. Never Before Heard Sounds is an AI-powered music production company, working to create new sounds and new musical datasets. SyncFloor is a marketplace of commercial music that can be used in movies, TV shows, ads, video games and elsewhere. The Next Big Idea Club offers a subscription for curated nonfiction books — you can buy the books themselves, but also read, watch or listen to condensed summaries. Here are the six startups in Betaworks’ new Audiocamp  

Read More...
posted 6 days ago on techcrunch
Google is giving an A.I. upgrade to its Collections feature — basically Google’s own take on Pinterest, but built into Google Search. Originally a name given to organizing images, the Collections feature that launched in 2018 let you save any type of search result — images, bookmarks, or maps locations — into groups called “Collections” for later perusal. Starting today, Google will make suggestions about items you can add to Collections based on your Search history across specific activities like cooking, shopping or hobbies. The idea here is that people often use Google for research but don’t remember to save web pages for easy retrieval. That leads to users to dig through their Google Search History in an effort to find the lost page. Google believes that A.I. smarts can improve the process, by helping users to build reference collections by starting the process for them. Here’s how it works. After you’ve visited pages on Google Search in the Google app or on the mobile web, Google will group together similar pages related to things like cooking, shopping, and hobbies then prompt you to save them to suggested Collections. For example, after an evening of scouring the web for recipes, Google may share a suggested Collection with you titled “Dinner Party” which is auto-populated with relevant pages from your Search history. You can uncheck any recipes that don’t belong and rename the collection from “Dinner Party” to something else of your choosing, if you want. You then tap the “Create” button to turn this selection from your Search history into a Collection. These Collections can be found later in the Collections tab in the Google app or through the Google.com side menu on the mobile web. There is an option to turn off this feature in Settings, but it’s enabled by default. The Pinterest-like feature aims to keep Google users from venturing off Google sites to other places where they can save and organize things they’re interested in — whether that’s a list of recipes they want to add to a pinboard on Pinterest or a list of clothing they want to add to a wish list on Amazon. In particular, retaining e-commerce shoppers from leaving Google for Amazon is something the company is heavily focused on these days. The company recently rolled out a big revamp of its Google Shopping vertical and just this month launched a way to shop directly from search results. The issue with sites like Pinterest is that they’re capturing shoppers at an earlier stage in the buying process — during the information-gathering and inspiration-seeking research stage, that is. By saving links to Pinterest’s pinboards, shoppers ready to make a purchase are bypassing Google (and its advertisers) to check out directly with retailers. Meanwhile, Google is simultaneously losing traffic to Amazon, which now surpasses Google for product searches. Even Instagram, of all places, has become a rival, as it’s now a place to shop. The app’s Shopping feature is funneling users right from its visual ads to a checkout page in the app. PayPal, catching wind of this trend, recently spent $4 billion to buy Honey in order to capture shoppers earlier in their journey. For users, Google Collections is just about encouraging you to put your searches into groups for later access. But for Google, it’s also about getting people to shop on Google and stay on Google, no matter what they’re researching. Suggested Collections may lure you in as an easy way to organize recipes, but ultimately this feature will be about getting users to develop a habit of saving their searches to Google — and particularly their product searches. Once you have a Collection set up, Google can point you to other related items, including websites, images,  and more. Most importantly, this will serve as a new way to get users to perform more product searches, too, as it can send users to other product pages without the user having to type in an explicit search query. The update also comes with an often-requested collaboration feature, which means you can now share a collection with others for either viewing or editing. Sharing and related content suggestions are live worldwide. The A.I.-powered suggested collections are live in the U.S. for English users starting today and will reach more markets in time.

Read More...
posted 6 days ago on techcrunch
  As we learned back in October, Microsoft has been cracking away at not one, but two dual screen devices: Surface Duo, and Surface Neo. Surface Duo will run Android, while the Surface Neo will run on a special fork of Windows 10 dubbed “Windows 10 X”. This morning the company is pulling back the curtain a bit, debuting its first batch of dual-screen developer tools and shedding some light on how apps can utilize that second screen. While a developer kit for the Android-powered Duo has been made available immediately, the company says dev tools for the Windows-powered Neo will arrive in “the coming weeks”, with a target date of February 11th. By default, says Microsoft, apps on these dual-screen devices will only occupy one screen. Users can elect to “span” the app to make it stretch across both — but, at least for now, it’s not something an app can force to happen. While simply stretching an app to fill both screens is one approach, Microsoft offered up a few alternative “pattern ideas” to better utilize the form factor: Meanwhile, Microsoft is also starting to build out web standards for dual-screen devices — APIs for developers to easily detect dual-screen devices, for example, allowing them to adapt their web apps accordingly. The company says preview builds of Microsoft Edge with early dual-screen APIs should start shipping “soon”. While no specific launch date has been given for either device, Microsoft has given both a launch window of sometime around the Holidays of 2020. Getting these dev tools out sooner than later, then, makes sense — while it might look neat, two screens aren’t inherently better than one. For the concept to ever take off, Microsoft needs developers to find the novel ways to use that second screen; the ways in which having a pair of screens really makes things better, rather than just… different.

Read More...
posted 6 days ago on techcrunch
Google Cloud today announced Secret Manager, a new tool that helps its users securely store their API keys, passwords, certificates and other data. With this, Google Cloud is giving its users a single tool to manage this kind of data and a centralized source of truth, something that even sophisticated enterprise organizations often lack. “Many applications require credentials to connect to a database, API keys to invoke a service, or certificates for authentication,” Google developer advocate Sath Vargo and product manager Matt Driscoll not in today’s announcement. “Managing and securing access to these secrets is often complicated by secret sprawl, poor visibility, or lack of integrations.” With Berglas, Google already offered an open-source command-line tool for managing secrets. Secret Manager and Berglas will play well together and users will be able to move their secrets from the open-source tool into Secret Manager and use Berglas to create and access secrets from the cloud-based tool as well. With KMS, Google also offers a fully managed key management system (as do Google Cloud’s competitors). The two tools are very much complementary. As Google notes, KMS does not actually store the secrets — it encrypts the secrets you store elsewhere. The secret Manager provides a way to easily store (and manage) these secrets in Google Cloud. Secret Manager includes the necessary tools for managing secret versions and audit logging, for example. Secrets in Secret Manager are also project-based global resources, the company stresses, while competing tools often feature manage secrets on a regional basis. The new tool is now in beta and available to all Google Cloud customers.

Read More...
posted 6 days ago on techcrunch
International Business Machines is living a case study of a large, established company vying to transform. Over the last decade, the technology elder has struggled to move into areas like cloud and AI. IBM has leaned on a combination of its own R&D abilities and deep pockets to push into modern markets, but has struggled to turn them into revenue growth. At one point, Big Blue posted 22 sequential quarters of falling revenue, a mind-boggling testament to how hard it can be to turn around a juggernaut. More recently, IBM shrank for another five consecutive quarters, a streak it broke with yesterday’s news that it had beat analyst expectations.  The quarter brought modest, but welcome revenue growth. Perhaps more importantly, the company’s top line expansion was co-led by the old IBM mainframe business and its newest champion, Red Hat. IBM can be happy for the positive financial news, for now at least, but it needs to repeat the result. The challenge it faces moving forward will include finding a way to continue revenue growth while modernizing its product line and ensuring that its huge Red Hat purchase continues to perform.

Read More...